对现有的一些Vague集(值)相似度量算法进行分析,指出了这些方法在计算相似度方面的片面性与局限性。通过对Vague集中元素的未知度进行分析,并引入支持系数来调节未知信息对真、假隶属度的影响,从而提出了新的Vague集(值)相似度量算法,通过模式匹配的实例验证了该算法的实用性。通过与以往的一些算法进行比较,充分证明了新算法的合理性与有效性。
This paper analyzes some existing methods of similarity measures and pointes out the unilateralism and localization of these methods in calculating the similarity measures between Vague(values) sets.Through analyzing the hesitancy degree of the elements in the Vague(values) sets and introducing the upholding coefficient,which is used to adjust the influence of the unknown information to the true and the false subjection measures, a new kind of similarity measure is educed in the Vague(values) sets,and it is proved usefully by the example of the pattern matching.The rationality and validity are proved by comparing the new method with the others.